Wired’s Infoporn on mapping the price of a calorie
The current issue of Wired (16.01) has an infographic feature I put together, with three visuals overlaying the price per calorie of various foodstuffs with their energy payoff and sugar content. Basically, it shows how the cheapest foods in a supermarket are generally the most likely to make you fat. This won't be news to most of you, but I was pleased to do it for a magazine like Wired, which tends to be a pro-biotech, anti-hippie-foodie publication with a wide reach. Believe it or not, it wasn't my idea — the research editor asked me if I was willing and of course I said yes.
Comfort me with apple pies
Research by University of Washington nutritional science prof Adam Drewnowski indicates that the highest obesity rates are found among the least educated and poorest members of society. (See his 2004 paper for the American Journal of Clinical Nutrition, "Poverty and obesity: the role of energy density and energy costs.") These are the people most likely to be looking at price tags, not nutritional labels, at the supermarket. And there's the problem. Drewnowski has found that products supplying the cheapest calories also tend to offer the most "energy density," that is, calories by weight — meaning it takes less to fill up a hungry stomach. That's because energy-dense foods are usually low in water but high in fat and sugar, two elements we are evolutionarily predisposed to seek out, and the same two ingredients that are artificially cheap for processed-food manufacturers, thanks to a complex system of government subsidies and tariffs. Here's the real bottom line: Eat enough "comfort food," and your body will get more comfortable, until you're literally a walking La-Z-Boy.
I went on two fact-finding missions to a local Safeway to document the cost of 78 food items on a per-calorie basis, photographing the shelf pricetags and the nutritional labels surreptitiously. (My Lumix has a macro setting that allowed me to basically palm the camera, hold it up a few inches from the item, and shoot without anyone being the wiser.) Then I compared how many calories were found in a 100-gram serving to determine energy density. Sure, you're not going to eat 100 grams of Newman's Own ranch dressing, but nor are you going to stop at 100 grams of Cherry Garcia. Recommended serving sizes are useless on many levels, including for comparing energy density. Then I checked labels and the USDA's database of food composition to learn how many grams of fat, sugar, and fiber were in those 100 grams. Sure enough, the cheapest foods were also the most processed and the worst for your waistline (high in fat and sugar, low in fiber).
The good food/bad food dilemma
Gathering the data was the easy part. The challenge was how to represent visually which foods were healthy and which were not. The Wired editor wanted me to be as scientific as possible. Well as you all know — or will soon, once you read Michael Pollan's new book — nutrition science is constantly in flux. Fats are bad! No, fats are OK — carbs are bad! Butter! No, margarine! She suggested I ask New York University nutritionist Marion Nestle if we could apply a formula such as 2 points for every gram of protein or fiber per serving, minus 2 points for every gram of sugar, and so forth.
I winced, but I asked. Unsurprisingly, Nestle's answer — sent from her Blackberry in Delhi — was unequivocal: "I am completely opposed to these sorts of schemes because they are a slippery slope. Lots of groups have tried this. Hannaford's is probably the best. CSPI also has one. None of them work perfectly. I much prefer food-based approaches."
However, she did remind me of Drewnowski's work, which she had cited in her 2006 book "What to Eat," and energy density turned out to be the key to unlocking my massive spreadsheet. The cheapest calories offer by far the most energy bang for a shopper's buck mainly because they are full of sugar — and fat and salt, although that's not in the graphic — more than anything else. Here are the five most expensive and cheapest foods from my highly unscientific survey, along with their energy density and sugar content rankings.
| Food item | Energy cost ($ per calorie) |
Energy density (calories per 100 grams) | Grams of sugar per 100 grams | |
| 1 | Tomatoes, grape | $0.07825 | 0 | 0.0 |
| 2 | Green leaf lettuce | $0.02922 | 18 | 2.6 |
| 3 | Stouffer's Lean Cuisine Everyday Favorite Roasted Turkey and Vegetable | $0.02919 | 15 | 0.8 |
| 4 | Harley Davidson Beef Jerky Teriyaki flavor | $0.02277 | 66 | 1.8 |
| 5 | Pineapple spears, 16 oz | $0.01761 | 286 | 7.1 |
| 74 | Butcher's Cut Jumbo Franks | $0.00104 | 316 | 7.0 |
| 75 | Duncan Hines Frosting | $0.00104 | 429 | 45.7 |
| 76 | Mission Corn Tortillas 36 Ct | $0.00079 | 216 | 3.9 |
| 77 | Maruchan Ramen Noodle Soup Pork Flavor | $0.00067 | 442 | 0.0 |
| 78 | South Beach Diet High-Protein Cereal Bars - Cranberry Almond | $0.00062 | 400 | 20.0 |
As it happens, Drewnowski and his researchers one-upped Wired a few weeks ago by releasing a new study of 370 foods at a Seattle supermarket. As the New York Times coverage of the report said, not only do junk foods cost a fraction of the price of fruits and vegetables — energy-dense snackfoods average $1.76 per 1,000 calories, compared with $18.16 for lower-energy but nutritious foods — but junk food prices also are more immune to inflation.
No related posts.



Comments
By Kim on December 31st, 2007 at 1:29 pm
In fact, I saw your map in Wired just this morning as I ate my breakfast -- nice job. Thanks for the backstory!
By Amanda Rose on January 3rd, 2008 at 10:27 am
Super duper. Wired is one of my husband's favorite magazines. Your chart must be here somewhere. :)